In the past 15 years, neurobiologists have challenged the view that memories stabilize over time through a process of consolidation; recent research indicates that memories undergo a period of restabilization following reactivation/retrieval that i distinct from initial consolidation. Termed reconsolidation, this process appears to be evolutionarily conserved and dynamic - maintaining, strengthening, and/or updating existing memories in order to preserve their predictive relevance. Reconsolidation is of interest to researchers and clinicians, in part because it holds tremendous promise for those who suffer from intrusive and often debilitating memories, such as patients with post-traumatic stress disorder (PTSD). Reconsolidation research has the potential to inform and facilitate cognitive-behavioral therapy, resulting in better outcomes for PTSD sufferers. Unfortunately, the basic behavioral and neurobiological mechanisms of reconsolidation are poorly understood. However, preliminary investigations by our lab and others suggest an important role for prediction error (i.e., surprise, or the violation of expectation), in the initiation of memory reconsolidation processes. If reconsolidation provides a mechanism for the updating of existing memories, then new (i.e., not already predicted) information must be present in order for reconsolidation processes to be initiated. Using a combination of behavioral, anatomical and pharmacological approaches, we propose a systematic analysis of the neural circuits responsible for the prediction error signal that triggers reconsolidation of contextual fear memories. In Aim 1, we wil examine contextual fear memory reconsolidation following an unexpected retrieval. We hypothesize that reactivation of a contextual fear memory in an unexpected manner will initiate fear memory reconsolidation processes in the basolateral amygdala (BLA). Either the omission of an expected outcome or the substitution of an expected outcome with a different aversive stimulus will initiate reconsolidation. Further, we hypothesize that BLA expression of both zif268 and Arc/Arg3.1 will be elevated following an unexpected reactivation session. In Aim 2, we will define the role of ventrolateral periaqueductal gray (vlPAG) opioid receptors in triggering contextual fear memory reconsolidation following an unexpected retrieval. We hypothesize that vlPAG opioid receptors trigger amygdalar reconsolidation of fear memories during memory reactivation that violates an expectation. In Aim 3, we will define the role of the orbitofrontal cortex (OFC) in triggering contextual fear memory reconsolidation following an unexpected retrieval. We hypothesize that the OFC will play a critical role in signaling outcome expectancy and thereby triggering BLA reconsolidation processes when the reactivation session violates an expected outcome. Our application represents a major conceptual innovation, as it is the first to look at reconsolidation as the product of interacting components of an integrated error-correction circuit.